Knee arthrodesis is most commonly performed for failed total knee arthroplasty. Conventional arthrodesis techniques are associated with a high incidence of complications and are unsuitable in cases with extensive bone loss. We report our medium-term results using a custom-made cemented knee arthrodesis prosthesis in 10 patients with a mean follow-up of 56.4 months (range, 15-199 months). The prosthesis was implanted as a 1- or 2-stage procedure for infected revision knee arthroplasty or tumor endoprosthesis in 9 patients and as a primary procedure in 1 patient with angiosarcoma involving the knee extensor mechanism. The average combined femoral and tibial bone deficit was 170 mm (range, 56-220 mm). Implant survivorship was 90%. All patients with retained prosthesis had no evidence of residual infection or loosening and were able to mobilize independently. One prosthesis was revised though retained following a prosthetic fracture, and 1 patient underwent above-knee amputation for uncontrolled infection. We conclude that the Stanmore knee arthrodesis prosthesis provides reliable fusion in an otherwise difficult-to-treat group of patients.